Here are the steps of how this happens, as I see it, though not every one of these steps occurs in every single case:

(1) Many people feel dissatisfied with the state of their society and its impact on their lives, and blame the groups that have long remained in power. (Sometimes these groups that have long been in power deserve blame for the bad state of affairs; other times, the bad state of affairs is due to circumstances largely outside of their control.) Blaming these groups that have long remained in power often also involves distrust in existing institutions.

(2) A new potential leader emerges who is perceived not to be a member of the groups that have historically been in power (or if they come from the original groups of power, then this potential leader typically claims outsider status, presenting themselves as a reformer who is different than the other insiders). Typically, this potential leader is viewed as being more “one of the people” than the prior groups in power.

(3) This leader is charismatic and, though potentially polarizing, many people find this person captivating and persuasive. Oftentimes, this leader is also high on traits commonly associated with grandiose narcissism (e.g., a belief in their own superiority and specialness, a need for admiration, arrogance, manipulative tendencies, a “with me or against me” mentality, and reduced empathy).

(4) This leader promotes a grand vision of what the country could become (or could be restored to again) under their leadership, which gets many people excited.

(5) The leader typically appears to be powerful and courageous, and willing to stand up against (and correct the errors and abuses of) the previous people in power who have held the country back.

(6) In addition to blaming society’s problems on the prior group in power, the leader also uses one or more other groups as scapegoats, and blames many of society’s problems on that group as well (typically a distrusted local minority group, immigrants, or some other country that the population sees as a potential threat). However, the scapegoat group(s) are either not actually a cause of any of the problems in that society, or are only a minor cause (with most of the problems being caused by other factors).

(7) The leader also exaggerates, lies to, and manipulates the people, claiming simple (but inaccurate) narratives about what’s gone wrong that make their solutions seem likely to be effective, and that paint themselves as the heroes and the prior groups in power (as well as the scapegoats) as villains.

(8) Through their charisma, exciting vision for the future, and promises to fix the problems in society by correcting the errors of the previous groups in power and by cracking down on the proposed scapegoats, the leader gets strong public support. The leader then gets into a position of executive power by one of a variety of means (e.g., being elected, being appointed, deal-making, coercion, or a coup).

(9) The leader then attempts to neutralize the power of formerly powerful groups that were in power (e.g., through violence, arrests, seizing their wealth, and/or disparaging them and damaging their reputations).

(10) The leader additionally harms the scapegoated group(s) (e.g., through arrests, preventing immigration, forced migration, tariffs, military action, and/or violence). Sometimes, though not always, the leader also cracks down on leading intellectuals and groups of intellectuals, attempting to weaken their influence or prevent them from being listened to.

(11) The leader amasses power and wealth for themselves while in charge, while also attempting to reduce the checks and balances on their power (e.g., by weakening or consolidating other parts of government, or by exerting control over media and journalism). Additionally, the leader typically engages in information control and propaganda in an attempt to get the population to view all of their actions as positive. The more effective their information control and propaganda are, the more they can get away with making terrible decisions on behalf of society and enriching themselves at the expense of the people.

(12) The leader’s actions sometimes produce no benefit at all for the bulk of the populace, and sometimes produce a short-term benefit to the populace, but ultimately, the leader’s actions come at a meaningful (and sometimes disastrous) long-term cost, leaving society worse off long-term than it was before that person rose to power.

And, with that, as has happened again and again in many societies throughout the world, populations end up trying to improve their country’s problems by electing an exciting new leader, but instead end up being lied to and manipulated, leaving the well-being of the people of their country worse off than it would have been. This is a worldwide phenomenon. It happened with Hitler in Germany, Kim Il-Sung in North Korea, Mao Zedong in China, Mugabe in Zimbabwe, and Putin in Russia. And it will likely keep happening. A first step towards helping stop this pattern is if voters become more aware of it.

To be clear, populism isn’t bad, per se. The unmet needs and desires of ordinary people who feel that their concerns are disregarded are very important. And widespread dissatisfaction with the status quo usually does mean that something needs to change. The real issue is that grandiose manipulators leverage populist sentiment to install themselves in power, leaving their people worse off in the long term.

It’s also important to observe that this pattern isn’t one of the right or the left – it’s a pattern that arises on both sides of the political spectrum.

This piece was first written on August 31, 2025, and first appeared on my website on October 23, 2025.

Sometimes managers as a group get a bad reputation. Some people even wonder whether managers are needed at all since they don’t seem to do any of the “real work.” There are also plenty of bad managers who actually impede the people they manage. And there’s the pretty common phenomenon of a company having too many managers, which leads to frustration and inefficiency.

But good managers do a number of very important things, even though it isn’t always obvious what all of these things are. I think that even pretty well-established managers sometimes wonder what it would look like to do their jobs better (I know that I do).

To help you remember what it means for someone to be a good manager, just think to yourself that “a good manager C.A.R.E.S.” In short, a good manager cares about their team working effectively together as an interdependent unit (Cohesion), cares about achieving the mission of the overall organization (Aim), cares about the rules and regulations their team needs to follow (Requirements), cares about the individual efficacy of their team members (Effectiveness), and cares about the happiness of the people they manage (Satisfaction).

Below is the full framework.

(1) Cohesion – help the team to work constructively together as a cohesive unit so that all the important tasks get done.

More specifically, this may require a manager to do things such as:

• Source Staff – make sure the team has enough staff members and make sure that everyone is in an appropriate role in order to accomplish the team’s goals

• Source Skills – make sure the members of the team have the various skills needed to accomplish the team’s mission, or else call in external help from those with the requisite skills

• Enhance Efficiency – find ways to accelerate the productivity of the team as a whole

• Clarify Responsibilities – make sure team members take full responsibility for, and are clear on, their respective responsibilities that affect the team’s success, and avoid having multiple people believing they are in charge of the same thing

• Prevent Oversights – make sure that someone has taken responsibility for each important part of the work that the team needs to do so that nothing slips through the cracks

• Clarify Norms – encourage and clarify the setting of both work and communication norms to set expectations about behaviors (e.g., “What time are people expected to go home by?” or “Is it acceptable to call work colleagues on Sunday for work-related non-emergencies?”)

• Ensure Communication – ensure good communication so that those who have valuable information share it with those who can benefit from it and so that when someone needs information that someone else on the team may have, they can get it

• Avoid Duplication – prevent duplication of work, and make people aware of related work done within the team and elsewhere in the organization, so that team members know what others are doing and what has been done before that may be relevant to them

• Resolve Disagreements – identify problems between team members, and help them figure out a solution (e.g., interpersonal conflict or disagreements about what to do)

• Focus Team – make sure the team is on the same page about the team’s near-term goals so that the output of different team members is aimed in the same direction

• Set Processes – put in place team and individual processes that encourage productivity, good communication, and high-quality output

• Fill Holes – if important work suddenly can’t get done by your team (e.g., because someone got sick, or because of an emergency, or because someone quit, or because of a miscommunication, etc.), jump in and do the necessary work yourself, or find someone who can immediately step in

• Improve Processes – figure out how the existing work processes could be improved or made more efficient

• Promote Safety – help ensure that team members feel the psychological safety to be vulnerable and honest in front of each other, to put forth ideas that might turn out to be bad ideas, to ask questions that might come across as naive, to admit mistakes, etc.

• Set Deadlines – create non-arbitrary deadlines and timelines that help keep the whole team focused on efficiently achieving milestones together

(2) Aim – ensure that the team’s output moves the organization as a whole towards the organization’s overarching goals.

• Maintain Vision – make sure the team is producing work that targets the vision of the overall organization, and remind the team of the vision periodically when the team’s output drifts off course

• Set Goals – work with the team to set team level goals and milestones

• Make Plans – work with the team to form plans for how to efficiently achieve the team’s long-term goals

• Prioritize – help team members prioritize so that they are doing what is most useful for ultimately achieving the organization’s objectives

• Provide R.O.I. (return on investment) – make sure the team is producing a high “expected value” when it comes to the return on investment the team provides for the whole organization

• Promote Collaboration – make sure that the team cooperates with other teams in order to help other teams at the organization achieve their goals

• Decide – thoughtfully make critical decisions on behalf of the team about how a project should proceed, or help guide the team towards making their own well-informed decision (e.g., about which technology is best to use for a project)

(3) Requirements – make sure that the team doesn’t violate important legal, ethical, project and resource constraints.

• Follow Laws – make sure team members don’t break any laws

• Follow Rules – make sure team members don’t violate any important company policies

• Ensure Ethics – make sure team members don’t behave unethically

• Ensure Safety – make sure that work is carried out in a safe manner

• Handle Violations – intervene immediately if laws are broken, rules are broken, or unethical behavior occurs

• Get Resources – make sure that the team has the resources they need to succeed, for instance, by communicating with higher up management on behalf of the team

• Limit Resources – make sure the team doesn’t use more than its allotted resources (e.g., money or company equipment)

• Maintain Brand – Make sure the team doesn’t take actions that reflect badly on the company or organization’s image or brand; leave a positive impression on customers, journalists, and the public

• Avoid Toes – make sure the team doesn’t undermine or create conflict with other teams at the organization

• Fix Rules – if company-wide rules, organizational structure, or processes are preventing the team from working effectively, try to get that structure modified or the rules changed

(4) Effectiveness – help each individual team member be effective at carrying out their own independent work.

• Grow Skills – help team members figure out how to improve their skills

• Identify Weaknesses – make team members aware of their weaknesses in a way that doesn’t demotivate or offend them, and help them develop an effective strategy for improving

• Problem Solve – help team members problem solve any challenges that come up in their work and help them get unstuck if they aren’t sure what to do, can’t seem to get through a challenging problem, or can’t seem to make forward progress

• Track Performance – keep track of how each team member is performing, and if any team member is not performing sufficiently well, work with them to increase their performance; only fire them as a last resort

• Incentivize – incentivize good work, and disincentivize bad work

• Avoid Blocking – don’t slow down or delay the work of team members without good reason, and don’t keep team members waiting a long time for your feedback or help

• Give Feedback – give feedback on work, both as a way to help make that specific work output and as a way to help the team member improve similar work in the future

• Evaluate Quality – examine the work output of individual team members to evaluate what sort of work they are good at now (that they can do with little or no supervision), what requires supervision, what should be handed off to someone else who has stronger relevant skills, and what new work this person should be given

• Enhance Productivity – figuring out ways to help make team members more productive, whether through changes in the environment, the processes used, the equipment available, the team or communication structures, etc.

• Navigate Bureaucracy – help team members navigate the organization’s systems and bureaucracy to make sure they get the resources and information they need to do their job well

• Shield Team – protect the team from distractions and from low-value or unreasonable demands coming from above

• Encourage Risks – help team members feel comfortable in taking smart (i.e., high expected value and limited in magnitude) risks

• Establish Goals – make sure that each team member has relevant goals for their work that are specific, challenging, and attainable, and that these individual goals are moving the team as a whole towards its overarching goals

(5) Satisfaction – help ensure that the team members are happy and that they don’t want to quit.

• Motivate – help the team members stay excited about the work or feel meaning in it, remind them of the bigger picture of why their work matters (to others or to themselves), taking into account their individual motivations (e.g., learning, financial security, solving challenging problems, helping other team members succeed, improving the lives of customers, self-expression, etc.)

• Praise – Make sure to let team members know when they do a good job (e.g., giving at least 90% positive feedback and less than 10% critical feedback, since negative feedback tends to have a substantially stronger impact)

• Give Wins – Give the team regular wins, and the sense of completing goals, rather than a never-ending feeling of trudging forward

• Reduce Stress – Make sure that stress or anxiety caused by the work is not more than team members can handle

• Resolve Conflicts – Make sure to resolve conflict with and between team members as quickly as possible

• Build Trust – Help create trust with your team members (e.g., they should know that they can confide in you and that you are looking out for their interests)

• Communicate Care – make it clear to team members that part of your job is to make sure they are happy at work and that they should come to you if they are not

• Identify Burnout – identify when employees are struggling, and when it happens, show empathy, give support, and help them get back to a point where they aren’t burned out

• Improve Self – get robust (preferably anonymized) critical feedback about how you could be a better manager (e.g., by having both your team members and your boss fill out an anonymous survey with suggestions for how you can improve)

• Dissect Problems – when problems are occurring with a team member, make sure to get to a root understanding of what is actually happening so that you can develop an effective plan for resolving it (rather than developing inadequate solutions based just on what appears to be happening at a superficial level)

• Staying Professional – keep one’s own (and other people’s) appearance and actions professional, and avoid behaviors that might offend or upset others

• Contain Emotions – do not let one’s own negative emotions bleed into interactions with team members, for instance, by having one’s own anxiety stress out other team members or having one’s own irritability or bad day create a bad day for others

• Get Help – make sure that if you yourself are struggling to do your job well as a manager, or if you are burning out, you get the needed resources, rest, or help

• Get Buy-in – make sure the team is on board with the team’s overall goals and plan

• Identify Unhappiness – identify who is unhappy or thinking of quitting, and work to find ways to make them happier

• Develop – help team members grow into new roles and responsibilities, and give them opportunities and resources to improve their skills

• Avoid Micromanaging – help make sure that team members get their ideal level of autonomy, striking a balance between giving too little and too much guidance

• Give Credit – when a person or team does a great job, make sure they get credit and recognition for that

• Shield – shield team members from blame when they don’t deserve it or when it isn’t helpful for them to be blamed

• Communicate Upward – make sure higher-ups understand how the team’s work fits into helping the organization achieve its goals so that the value of the team is understood throughout the organization

• Set Norms – set positive norms for how team members treat each other (e.g., people being kind, respectful, helping each other, etc.)

• Communicate Impact – make sure the team is aware of the positive impact that their work has had on customers, other teams, and the organization as a whole

A SAILING METAPHOR

Here’s a sailing metaphor that I think helps make this C.A.R.E.S. categorization more intuitive.

Think of an organization as a squadron of sailboats. They are aiming to get to a particular place together, with each boat carrying a different essential cargo. The manager is the captain of one of these sailboats, and the manager’s team is the crew of that boat.

With this sailing metaphor in mind, here are the categories again, briefly:

Cohesion: this means making sure that the crew works together to raise the sails when the sails are needed. When rowing must be done, it means making sure the crew’s oar strokes are synchronized so that the boat goes as fast as possible. And it means making sure that every oar available in the boat is being rowed by someone.

Aim: this means the captain is making sure the boat is going in the same direction as all the other boats in the squadron. That way, the squadron as a whole can achieve its overall goal of getting to where it is going. This involves the captain communicating with their own crew, as well as with the captains of the other boats.

Requirements: this might mean, for example, making sure that the boat doesn’t go too close to protected coral reefs and making sure the boat doesn’t have more oars on it than it needs (that could be useful to use in other boats). It also might mean making sure that nobody smokes on the boat (which could be a safety hazard due to potential fire).

Effectiveness: this would mean making sure each individual team member is good at handling their particular duties on the boat, for instance, knowing how to tie strong knots and using proper form when rowing. And it means making sure that if an oar breaks, the team member it belonged to gets a new oar right away.

Satisfaction: this might involve reminding the team of the importance of the squadron’s overall mission and of their own boat’s critical role in that mission. And it means making sure that the crew is happy and that nobody wants to abandon ship.

Putting this all together: if the crew is working together cohesively, the boat is aimed in the right direction, the crew members don’t violate any important requirements, each crew member is doing their own job effectively, and everyone is satisfied enough to see the journey through, then the boat should succeed in doing its part to help the entire squadron get to where it’s going.

This essay was first written on October 22, 2019, and first appeared on this site on April 15, 2022.

1. Looping the Question

“What don’t I know that I must know?” This Meta Question is your obsession. The answer to that Meta Question is itself a question, and it leads you to create your first data loop (say, conducting a series of interviews with experts). After sucking down much of what it has to teach you, you’re hearing the same answers again and again. Data loops are ephemeral, but this one has served its purpose.

“What don’t I know that I must know?” you ask again. The answer yields your next question, which leads to your next data loop (say, reading many articles). Then that stops being as useful, and you construct your next loop (e.g., conducting a survey) and another (e.g., watching people use your product) and another (e.g., getting criticism from UI experts) and another (e.g., studying the statistics of user behavior) and yet another (e.g., attempting to sell your product at different prices). For each question, you look to the many tools on your tool belt (e.g., user interviews, expert feedback, surveys, A/B tests, analysis of user behavior, etc.) to pick the tool that is most suited for answering it. You are a master asker of questions, an expert looper of the answers.

2. Programming the Machine

You are constructing a machine out of gears. Some parts are made of metal or plastic, but mostly they are made of bits or people. Every month the machine produces a series of outputs, some intermediate, some final. You inspect these carefully.

Do the outputs have the right weight, color, clarity, cost efficiency, quality, quantity? Which part of the machine is bottlenecking the rest? Which part could use lubrication, or gas, or computation, or money, or inspiration? Which part of the machine is grinding to a halt, working against the rest, rotting inside, or about to fly apart? You perform intuitive calculus. What gear has the highest derivative of output with respect to dollars (or hours) of further investment?

As you feed it, the machine keeps growing bigger and more complex. You no longer have a complete understanding of how it operates – nobody does. It has ever more parts that can break, greater momentum which resists changing direction, more fuel needed to move it, more of a mind of its own. You battle the exploding complexity while the metal/plastic/flesh/binary monster of your creation lurches forward, smoke and steam spewing from its exhaust pipes. If you expertly wield the tools at your disposal, you can still usually manage to nudge it a tad to the left or a bit to the right, making sure it stays roughly on course.

3. Polishing the Stone

At first, you have a crude stone with countless rough edges. You show it to everyone. It looks like beach trash. It does almost nothing. People are confused why you are so excited about it. No one can see the form you see, trapped inside (though your friends try to, or at least try hard to pretend that they do). “Who would want that?” people silently wonder as they nod at what you’re saying. And these people are correct – no one wants it. You’re peddling rubbish.

But you believe in your stone. With enormous effort, you smooth the stone’s thousand rough edges. “You didn’t really need to smooth edge #732”, someone points out. “One little edge makes almost no difference.” This person is correct. Smoothing any single point makes *almost* no difference. Yet by smoothing one thousand rough edges, you’ve created a beautiful gem. It’s now the best of its kind in all the world. It’s now desired. “The idea was obvious,” people mutter, “I could have thought of that.”

4. Outlasting the Jungle

You’ve just parachuted into a dense jungle. Everyone knows there are treasures of incredible value buried there – in many different locations, in fact – but nobody knows where the treasure is. To make matters worse, the jungle is crawling with snakes, spiders, alligators, hungry bears, angry bears, crazy bears, bats, spike traps, people-eating plants, dragons, and buffalo, plus far too many other treasure hunters like you.

The vast majority of those who enter the jungle will end up fleeing it empty-handed. After a few snake/bear/buffalo/dragon bites and a few miles (or hundred miles) of strenuously hacking through the jungle, without ever quite knowing which direction to go, most have had enough.

Occasionally, though, people leave the jungle with some treasure in tow. And every once in a while, a group emerges with what is truly a king’s fortune. There is no shortage of treasure hunters. But you are different than the others. You are prepared, courageous, and confident – and you learn quickly. You have plenty of funding, a knife gripped between your teeth, a crossbow on your back, a net slung across one shoulder, bear spray for the bears, dragon spray for the dragons, grenades in case things get really tough, a tiny army at your back, 22 different maps written by 22 former adventurers (that semi-contradict each other), and a determination of tempered steel.

You know that if you can stay in the jungle longer than all the others – if, when others are falling down and coming apart, you can rally and try ten more paths to glory – while gaining skills from each failed outing – you will vastly increase your chances of success.

But then again, as you look around, you see dense thickets full of thorns in all directions. The tree cover blots out most of the sun, casting long, foreboding shadows. There is a faint bellowing roar in the distance, but the pitch is too low to identify its direction. And sweat beads on your forehead. “Which way?” your tiny army asks you.

So you’re thinking of starting a startup? Become a looper of questions to know how to program the machine that will polish the stone while you outlast the jungle.

This essay was first written on December 22, 2018, and first appeared on this site on January 28, 2022.

Thinking along these lines, each of us can divide all things that exist or seem to exist into a few categories:

1. Those things that exist in physical reality that also have a representation in my simulation. An example of this would be some particular car that I see on the street. There are atoms (or quantum wave functions at least) in the universe that correspond to this car, and there is a visual experience in my simulation that represents this car. After I look at this car and walk away, my simulation continues to have the concept of this car, at least until I have forgotten about it. So the car exists in my simulation, but it also exists in physical reality as a bunch of atoms (though there is no dividing line in physical reality between where the car atoms end and not-car atoms begin, that dividing line too is a property of the simulation). Color is another thing in our simulation that corresponds to physical reality, but is of course only a representation of it. There is no “redness” in space, redness is something that we experience. What there is in space is electromagnetic radiation, some of which produces an experience of redness in our simulations after it hits the simulation/reality interface that we call our eyes.
2. Those things that exist in physical reality but have no representation in my simulation. This would include a particular planet in another galaxy that I’ve never heard of. Although I have the concept of planets existing, this particular planet has no representation in my simulation of reality.
3. Those things that exist in our simulation but are not representations of specific things in physical reality. An example of this would be a hallucination (sure, there are bunnies in physical reality, but that particular bunny that you think is laughing at you now and doing head stands doesn’t actually correspond to any set of atoms). Perhaps a more interesting example from this category though would be impurity. A dish is unclean, Jewish custom says, if it has touched both milk and meat. This contamination property continues to hold even after the plate is scrubbed completely clean, even if every atom of milk and meat are removed. Suppose that a man who lives by this system of thought is given a plate, and then he is told that it was recently used to eat both meat and milk. Regardless of whether that claim is true, so long as he believes it the plate will suddenly gain the property of uncleanness in his simulation. He may experience the sense of this uncleanness quite strongly, and the thought of eating from that plate may even cause him to feel disgust.

One thing that is odd about living inside a simulation (as we each are), is that aspects of the simulation that don’t correspond to anything in physical reality can feel just as real as those that do correspond. For instance, someone I know once had a schizophrenic period where they became convinced that certain objects in their kitchen were, in fact, quite evil. This evilness was completely convincing and quite overwhelming. But what does it mean for an object to be evil? Evilness surely is not a property of atoms, quantum wave functions, or physical objects, but a property of representations of things inside our simulations. Normally we only associate this property with intelligent beings, but when the simulation goes haywire it can get associated with inanimate things. Just as we think some particular clown nose is red (i.e. our simulation contains the experience of redness which our brain links to its concept of that clown nose), we could end up thinking that some particular bar of soap is evil (i.e. our simulation contains the experience of evilness which our brain links to its concept of that soap).

Influences: Michael Vassar